Elucidation of epigenetic mechanisms, which mediate aberrant gene expression during malignant transformation may hasten the development of novel cancer therapies. Our previous studies have demonstrated that via complementary mechanisms including DNA methylation, polycomb repressive complexes (PRC) 1 and 2, and noncoding RNAs, cigarette smoke induces stem-like phenotypes that coincide with progression to malignancy in normal human respiratory epithelial cells (NREC) and enhanced growth and metastatic potential of lung cancer cells. Despite extensive studies, the genetic as well as epigenetic mechanisms by which cigarette smoke initiates lung cancers, and the cell(s) of origin of these neoplasms have not been fully defined. It has been postulated that fully differentiated cells may de-differentiate or reprogram to committed progeny in response to tobacco carcinogens. Consistent with these findings, we have demonstrated epigenetic reprogramming in human respiratory epithelial cells following long term exposure to cigarette smoke. In addition, recent studies indicate that the human lung contains undifferentiated, self-renewing, pluripotent cells in niches within the distal airways. In order to better model pulmonary progenitor/stem cells and characterize epigenetic mechanisms contributing to pluripotency in lung cancers, we recently generated induced pluripotent stem cells (iPSC) from normal human small airway epithelial cells (SAEC) by lentiviral transduction of Yamanaka factors. The lung iPSC (Lu-iPSC) exhibited complex alterations in DNA methylation, with marked up-regulation of PRC-2-related genes, and modulation of 15,000 additional genes. Additional Sex Combs Like-3 (ASXL3), an epigenetic modifier not previously described in reprogrammed cells or any human malignancy, was up-regulated 400-fold in Lu-iPSC and was markedly over-expressed in SCLC lines and primary tumors. Knock-down of ASXL3 inhibited proliferation and teratoma formation by Lu-iPSC, and significantly diminished growth of SCLC cells in-vitro and in-vivo. Collectively, these studies highlight the potential utility of the Lu-iPSC model for elucidating epigenetic mechanisms contributing to pulmonary carcinogenesis, and identifying novel targets for lung cancer therapy. Limited information is available pertaining ASXL3 in normal or malignant cells. ASXL3 is thought to interact with EZH2 to recruit PRC2 to chromatin. Additionally, ASXL3 interacts with LSD1, possibly to maintain bivalent chromatin. Furthermore, ASXL3 interacts with BAP1, a nuclear deubiquitinase that reverses the repressive histone mark placed by PRC1. Because EZH2 and LSD1 are also up-regulated in subsets of SCLC, the effects of ASXL3 over-expression in these cancers may be context dependent. Our attempts to address this issue have been hampered by the lack of a reliable and specific antibody to ASXL3, and difficulties related to cloning and expression of ASXL3. We have now successfully recloned ASXL3 cDNA with a flag-tag into an expression vector so that we can perform mass spectroscopy experiments to identify novel binding partners, and to conduct definitive studies on the epigenomic, transcriptomic and phenotypic effects of ASXL3 expression in lung cancers. In collaboration with Dr. Gordon Hager's team, we have performed DNAse hypersensitivity sequencing (DHS) experiments with footprint analysis to identify transcription factors that mediate ASXL3 up-regulation in Lu-iPSC and SCLC. Additional DHS and RNA-seq experiments have been performed to identify genes which are critical for pluripotency in SCLC that possibly can be targeted for therapy of these neoplams. Results of these studies are being written up for publication. ASXL3 maps to a region of chromosome 18 that has not previously been identified as abnormal in lung cancers. Our CGH array experiments demonstrated that over-expression of ASXL3 in SCLC correlates with increased gene copy number. Interestingly, another poorly characterized gene which maps to the same region of chromosome 18 is also markedly up-regulated in Lu-iPSC and SCLC. Over-expression of this latter gene in SCLC is also highly correlated with increased gene copy number, suggesting that a region of chromosome 18 is amplified in SCLC. Experiments are in progress to examine this issue. The Lu-iPSC model provides a unique opportunity to examine epigenomic and phenotypic effects of cigarette smoke on respiratory stem cells. Although iPSC and embryonic stem cells are reported to be highly resistant to carcinogens, we have observed that cigarette smoke induces phenotypic alterations in Lu-iPSC that are suggestive of neuronal differentiation. If this is true, it could explain why there is far more transcriptomic overlap between SCLC and Lu-iPSC than NSCLC and Lu-iPSC. Presently, we are attempting to refine the exposure conditions for long term experiments; we anticipate that cigarette smoke will induce differentiation and eventual malignant transformation of Lu-iPSC, thereby allowing us to model the lung cancer stem cell. Related experiments are underway to comprehensively characterize epigenetic landscapes of established lung cancer lines sorted into pluripotent side population (SP) and non-SP fractions relative to Lu-iPSC cultured in the presence or absence of cigarette smoke. Collectively these efforts may enable us to define an epigenomic continuum from normal Lu-iPSC, to transformed Lu-iPSC, to SP, and then to non-SP fractions of lung cancers. In doing so, we expect to identify novel epigenetic mechanisms driving pluripotency that can be targeted for lung cancer therapy. Significant efforts are also underway to use iPSC technology to reprogram lung cancer cells to create induced pluripotent cancer cells (iPCC) for use as personalized vaccines targeting shared as well as unique tumor associated antigens for lung cancer therapy. iPCC will contain all of the neoantigens present in the primary lung cancers. Furthermore, reprogramming may restore expression of genes which regulate antigen processing and presentation that have been epigenetically silenced during malignant transformation or immuno-editing. We have recently reprogrammed SAEC from two additional donors for immune recognition experiments examining the feasibility of using Lu-iPSC as vaccines targeting cancer-testis antigens and endogenous retroviruses. These efforts are intended to directly translate to evaluation of iPCC and/or iPSC vaccines in lung cancer patients at the NIH Clinical Center. Our current reprogramming experiments represent novel, high-risk, high-reward collaborative efforts among established intramural and extramural investigators. It is anticipated that these studies will provide new insights pertaining to epigenetic mechanisms contributing to lung cancers, and facilitate clinical development of innovative epigenetic immunotherapies for these malignancies.